Introducing Eleven Realtime, a new class of Speech to Speech models.
Voice agents usually make you choose between quick replies or precise answers.
Eleven Realtime gives you both. It starts talking in half a second and gets spoken reasoning questions right.
The future of voice AI is one system optimized end-to-end for more natural interactions at lower latency.
Eleven Realtime v1 Max scores 98.8% on Big Bench Audio and starts speaking in 0.57s. No model on the Artificial Analysis leaderboard is both faster and more accurate.
Eleven Realtime v1 mini is optimized for natural conversations at low cost.
It handles the things that make a voice agent feel human, like knowing when you’ve finished speaking, recovering when you interrupt, and not stopping when you say “mm-hmm.”
This model is great for AI assistants, educational tutors, and interactive characters.
Eleven Realtime v1 max is optimized for handling more complex tasks like customer support and lead qualification.
It reasons through multi-step requests and calls tools to look things up and take action.
Both models are available in ElevenAgents today.
Select Eleven Realtime as the model, and deploy production-ready agents anywhere your customers are: phone lines, your website, WhatsApp, or your contact center platform.
Plus, you get the full platform features like knowledge bases, guardrails, testing, and monitoring.
Try both models today:
elevenlabs.io/realtime
Oct 20, 2026 · 2:25 PM UTC